From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Van Haaren, Harry" <harry.van.haaren@intel.com>
Subject: Re: [PATCH v3 04/17] eventdev: add APIs for extended
	stats
Date: Mon, 20 Feb 2017 12:12:35 +0000
Message-ID: <E923DB57A917B54B9182A2E928D00FA6129F37B9@IRSMSX102.ger.corp.intel.com>
References: <1485879273-86228-1-git-send-email-harry.van.haaren@intel.com>
 <1487343252-16092-1-git-send-email-harry.van.haaren@intel.com>
 <1487343252-16092-5-git-send-email-harry.van.haaren@intel.com>
 <20170219123224.GC7400@localhost.localdomain>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Cc: "dev@dpdk.org" <dev@dpdk.org>, "Richardson, Bruce"
 <bruce.richardson@intel.com>
To: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mga07.intel.com (mga07.intel.com [134.134.136.100])
 by dpdk.org (Postfix) with ESMTP id 10AE52C66
 for <dev@dpdk.org>; Mon, 20 Feb 2017 13:12:38 +0100 (CET)
In-Reply-To: <20170219123224.GC7400@localhost.localdomain>
Content-Language: en-US
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

> -----Original Message-----
> From: Jerin Jacob [mailto:jerin.jacob@caviumnetworks.com]
> Sent: Sunday, February 19, 2017 12:32 PM
> To: Van Haaren, Harry <harry.van.haaren@intel.com>
> Cc: dev@dpdk.org; Richardson, Bruce <bruce.richardson@intel.com>
> Subject: Re: [PATCH v3 04/17] eventdev: add APIs for extended stats

<snip>

> > +/**
> > + * Reset the values of the xstats on the whole device.
> > + *
> > + * @param dev_id
> > + *   The identifier of the device
> > + * @return
> > + *   - zero: successfully reset the statistics to zero
> > + *   - negative value: -EINVAL invalid dev_id, -ENOTSUP if not support=
ed.
> > + */
> > +int
> > +rte_event_dev_xstats_reset(uint8_t dev_id);
>=20
> I think it would be useful to selectively reset the specific counter if
> needed.


Ok - I like the simplicity of a single reset() covering the whole eventdev,=
 so that the stats should be consistent. No objection to changing the API, =
as applications can do a full reset using the "partial" reset API if they r=
equire.


> something like below(Just to share my thought in C code)
>=20
> int
> rte_event_dev_xstats_reset(uint8_t dev_id,
> 	enum rte_event_dev_xstats_mode mode/* INT_MAX to specify all xstats on t=
he whole device */
> 	const unsigned int ids /* UINT_MAX to specify all the ids on the specifi=
c mode */

=09
The other functions that take a mode require a "queue_port_id" to select th=
e component (port number or queue id number). I'll add the parameter.

Also I don't like the INT_MAX solution, as the enum type doesn't have to be=
 INT, it can be char or uintX_t - compiler and CFLAGS can be used to change=
 the enum type IIRC.

The ids parameter is an array in the xstats_get() functions, allowing multi=
ple ids to be retrieved in one call. This also allows a NULL parameter to i=
ndicate all. I think this suits better than a single int id, and UINT_MAX f=
or all.


TL;DR: I'm suggesting

int
rte_event_dev_xstats_reset(uint8_t dev_id,
        enum rte_event_dev_xstats_mode mode,   /* device, port or queue */
        int16_t queue_port_id,                 /* Queue or Port to reset. -=
1 resets all of *mode* ports or queues. */
        const uint32_t ids[]);                 /* NULL array indicates to r=
eset all stats */

Thoughts?


> Other than above comments, Its looks very good.

Thanks for review!